Graduate Employment Prediction with Bias

Abstract

The failure of landing a job for college students could cause serious social consequences such as drunkenness and suicide. In addition to academic performance, unconscious biases can become one key obstacle for hunting jobs for graduating students. Thus, it is necessary to understand these unconscious biases so that we can help these students at an early stage with more personalized intervention. In this paper, we develop a framework, i.e., MAYA (Multi-mAjor emploYment stAtus) to predict students' employment status while considering biases. The framework consists of four major components. Firstly, we solve the heterogeneity of student courses by embedding academic performance into a unified space. Then, we apply a generative adversarial network (GAN) to overcome the class imbalance problem. Thirdly, we adopt Long Short-Term Memory (LSTM) with a novel dropout mechanism to comprehensively capture sequential information among semesters. Finally, we design a bias-based regularization to capture the job market biases. We conduct extensive experiments on a large-scale educational dataset and the results demonstrate the effectiveness of our prediction framework.

Prediction of College Students’ Employment Rate Based on Gray System

“…Some scholars proposed a gray system based on the gray theory, assuming that the research object has a certain gray color, analyzing it according to the current information known about the object, establishing a corresponding mathematical model, and then analyzing and quantifying the association between the factors of the object according to the model to get a model that can describe the change characteristics of the object [12]. In this section, we will discuss a common model of the gray system.…”

“…(1) Collecting historical data of the college students' employment rate of a university in recent years and eliminating some wrong data through experts to ensure the validity of the data. (2) Using equation (12) to process the original college student employment rate data to obtain better college student employment rate prediction results,…”

Graph Learning: A Survey

“…These vectors can be regarded as the features of input in previous machine learning methods. Graph learning methods pave the way for graph analysis in the new representation space, and many graph analytical tasks, such as link prediction, recommendation and classification, can be solved efficiently [10], [11]. Graphical network representation sheds light on various aspects of social life, such as communication patterns, community structure, and information diffusion [12], [13].…”

SUMMER: Bias-aware Prediction of Graduate Employment Based on Educational Big Data

“…Finally, we design a regularization to smooth the weight caused employment biases of diferent majors. This work is based on our previous work [24], which aimed to explore the bias of graduate employment and make an efective prediction for graduate employment. In previous work [24], we focused on predicting employment status only.…”

“…This work is based on our previous work [24], which aimed to explore the bias of graduate employment and make an efective prediction for graduate employment. In previous work [24], we focused on predicting employment status only. In the current work, we make prediction for employment preference in addition to ) [16].…”